JSON dumps no longer contain "type" entry in "files"

[bitsgalore]

The JSON dumps generated from the REST API no longer appear to contain file type info for individual files. See as an example the following publication:

https://zenodo.org/records/3529327

An old (late 2022) JSON dump shows the following "files" entry for this record:

"files": [{"bucket": "db947caf-b9ef-4d54-a3b2-489f41b12dfa", "checksum": "md5:d8f5af9e6cfa462b23f24b4718cc3461", "key": "Jpylyzer-trailer.mp4", "links": {"self": "https://zenodo.org/api/files/db947caf-b9ef-4d54-a3b2-489f41b12dfa/Jpylyzer-trailer.mp4"}, "size": 98680838, "type": "mp4"}]

When I created a new dump today (using the same API call), the "files" entry for the same record looks like this:

"files": [{"id": "f2aa6a1b-394a-4206-ab6a-e952e53de85d", "key": "Jpylyzer-trailer.mp4", "size": 98680838, "checksum": "md5:d8f5af9e6cfa462b23f24b4718cc3461", "links": {"self": "https://zenodo.org/api/records/3529327/files/Jpylyzer-trailer.mp4/content"}}]

Note the absence of the "type" entry in the new format! I used these dumps in a script that generates automated reports of our Zenodo community, which includes file type statistics that were based on the "type" value. This is no longer possible. It's not entirely clear to me if this is a bug or an intentional change after the recent upgrade.

(BTW I'm using this script to fetch the JSON dumps, which is just a straight adaptation of the docs)